The Performance of a Full Face Tunnel Boring Machine (TBM) in Tarabya (Istanbul)

The performance of a full face tunnel boring machine (TBM) in Tarabya (Istanbul)

N. Bilgin, C. Feridunoglu & D. Tumac Istanbul Technical University, Mining Engineering Department, Istanbul, Turkey M. Cnar, Y. Palakci, O. Gunduz & L. Ozyol Tinsa Co., Istanbul, Turkey

ABSTRACT: The performance of a full face tunnel boring machine (Herrenknecht TBM) was evaluated within the research program of this study. TBM having a diameter of 2.9 m and cutting power of 560 kW was used to excavate the tunnel of 13270 m in length, which was situated between Sariyer and Baltaliman. The main rock formations are Paleozoic aged sedimentary rocks, limestone, sandstone-siltstone, andesite dykes and sediment fillings. The project was commissioned to Tinsa – Öztas¸ – Hazinedaroglu – Simelko consortium. The performance of TBM was recorded continuously for detailed shift analysis. A limestone block having dimen- sions of 0.4 0.5 0.4 m, representing majority of the rock formations was collected from one of the shafts. This rock block was subjected to full scale instrumented laboratory cutting tests using a disc cutter taken from the TBM. The relationship between thrust force and depth of cut was compared to the one obtained in the field. It is concluded that accumulated data and instrumented full scale rock cutting tests may serve as a valuable tool to predict the performance of full face tunnel boring machines in any type of rock formation.

1INTRODUCTION

It is estimated that in the coming years the underground construction in Europe and in the other parts of the world will increase tremendously, typical exam- ples are tunnels, microtunnels, storage caverns (oil, gas, waste etc.). Mechanical rock excavation offers many advantages over conventional drill and blast techniques. These include improved safety, minimal ground dis- turbances, elimination of blast vibration, reduced venti- lation requirements etc. In addition, mechanical boring is capable of providing much higher rates of advance at lower cost than drill and blast excavation. A degree of uncertainty is always present on selecting equip- ment, predicting machine performance, estimating job duration, costs etc. However accumulated and pub- lished data from tunneling projects serve to improve predicting the machine performance and scheduling of job duration, this paper is prepared under this objective.

Figure 1. The general layout of the Tarabya Tunnel.

2 DESCRIPTION OF THE TUNNEL PROJECT AND TUNNEL GEOLOGY and Büyükdere bays around Istanbul Bosphorus. The tunnel having a length of 13,270 m and final diam- Tarabya tunnel is a part of sewerage project, which is eter of 2 m is situated between Saryer and Baltaliman planned to clean the sea pollution in Istinye – Tarabya as seen in Figure 1. 19 shafts are connected to the main

821 tunnel with 15 tunnels being oval in shape with a size 3 DESCRIPTION OF TBM AND WORK STUDY of 1200/1800 mm. Cross section of the main tunnel is ON MACHINE ASSEMBLAGE given in Figure 2. The tunnel mainly passes through limestone and shale rock formations of Silurian- General characteristics of Herrenknecht TBM used in Devonian age and sandstone, siltstone rock forma- tunnel drivage are given in Table 2. tions of Carboniferous age. Some magmatic intrusions TBM came to the field in 31 May 2000 and and sediment fillings are encountered in tunnel route. machine assemblage time took 30 days. Table 3 sum- Summary of the rock properties obtained from ten- maries machine assemblage time for different jobs. der document is given in Table 1. The figures in Table 1 A typical view of TBM is given in Figure 3. represent a tunnel length of 8847 m. The second part of the tunnel was also commissioned to the same con- tractors. However the second part of the tunnel was driven mainly in sediment fillings in majorities. Table 2. General characteristics of TBM used in Tarabya Tunnel. Some limestone samples were taken from the tunnel face in 18.09.2000 and 11.10.2000 for testing in the TBM excavation diameter 2915 mm laboratories. It is noted that the compressive strength TBM shield diameter 2870 mm of the rock samples taken from the face could be dif- Disc number 20 ferent than given in general documents. The compres- Disc diameter 305 mm (edge width 10 mm) sive strength values of the samples were 80 7MPa Max. rotational speed 16 rpm and 119 16 MPa. Max. torque applied 725 kNm Number of thrust cylinders 6 Max. thrust force 3750 kN (250 bar) Max power 620 kW

Table 3. Machine assemblage time for different jobs.

Preparation of the shaft bottom 6% Descending the cutting head and shield 7% Pushing cutting head and shield to face 7% Preparation of the rails for back-up system 3% Descending the back-up system 3% Preparation of the steel arch supporting the first rig 3% Assemblage of back-up and hydraulic system 25% Preparation of the conveyor band 3% Assemblage of the laser measuring system 7% Assemblage of the cooling system 7% Test running of TBM 11% Electrical breakdown 15% Figure 2. Cross section of the main tunnel. Cooling system breakdown 3%

Table 1. Summary of the rock properties encountered in tunnel route.

Rock Compressive Tensile Elastic formation % Strength strength modulus of the total (MPa) (MPa) (MPa) 103

Limestone 65% 44–81 4–5 9–15 Shale 17% 55–59 2.4 9–10 Sandstone- Siltstone 12% 59 – – Dykes 1% 32–40 3 6–7 Sediment Filling 5% – – – Figure 3. Typical view of TBM.

822 4 RECORDING THE PERFORMANCE OF TBM between 36 to 41%, which is a quite high compared to other applications. The machine utilization time was The tunnel excavation started in July 2000 and fin- found to change between 15% to 55% in general. ished in November 2004. The performance of TBM Machine utilization time is one of the most impor- was recorded continuously for detailed shift analysis. tant factors in determining the machine advance rate. Table 4 is a summary of overall performance for As seen from Figure 8, an increase in machine uti- 7700 m of tunnel drivage. lization time from 15% to 55%, increases the machine The performance of the TBM between chainage advance rate from 0.2 m/h to 0.8 m/h. Machine utiliza- 981 and 7700 m are summarized in Figures 4–7. As tion time is defined as the ratio of the net cutting time seen from these figures the average machine utiliza- of the machine to the total working time; i.e., shift time. tion time is 35%, breakdowns are between 24 to 27% of shift time. However it is interesting to note that the share of disc changing within the breakdown time is

Table 4. Overall TBM performance for 7700 m of tunnel drivage.

The best shift advance 18 m The mean shift advance 4.5 m The worst shift advance 0.75 m The best daily advance 24.75 m The mean daily advance 11.50 m The worst daily advance 0.75 m The best monthly advance 531 m Figure 6. Machine performance between chainage The mean monthly advance 376.7 m 2500–7700. The worst shift advance 144.75 m

Figure 7. Classification of breakdowns between chainage Figure 4. Machine performance between chainage 2500–7700. 981–2260.

0.9 0.8 y = 0.0143x - 0.01 2 0.7 R = 0.7078 0.6 0.5 0.4 0.3 0.2

Advance rate (m/h) 0.1 0 051015 20 25 30 35 40 45 50 55 60 65 Machine utilization (%)

Figure 5. Classification of breakdowns between chainage Figure 8. The relationship between machine advance rate 981–2260. and machine utilization time between chainage 981–2260.

823 5 FULL SCALE ROCK CUTTING TESTS acquisition card included eight independent channels. IN THE LABORATORY AND ESTIMATION Data recording rate is adjustable up to 50000 Hz. OF THE MACHINE CUTTING RATE The hydraulic cylinders can move the sample box in which the rock sample is cast with concrete to The main objective of the laboratory rock cutting tests eliminate pre-failure of the specimen. The cutter is was to see how the in situ measured machine perfor- fixed with a tool holder directly to the dynamometer. mance values were close to the predicted values from The specific energy obtained by dividing mean rolling laboratory rock cutting tests, which will contribute to force to yield (volume of rock removed per unit rock cutting mechanics for further industrial applica- length) is a key factor in determining the efficiency tions. For 4 mm depth of cut the measured net cutting of cutting process. It is advisable to work with opera- rate was 9 m3/h. Figure 9 represents the relationship tional parameters given minimum specific energy between disc thrust force and depth of cut measured which is obtained for a given cutter spacing/depth of in the field for 1 rev/min when excavating limestone cut ratio. formation. The philosophy behind the laboratory rock cutting Full-scale rock cutting tests were carried out in the tests lies in the reality that as a basic rule of rock laboratory using the cutting rig as outlined in Figure 10. cutting mechanics, for a given cutting tool and rock The rig was designed and manufactured within NATO- formation, optimum specific energy is obtained for a TU excavation project. A high quality aircraft alu- defined s/d ratio. Specific energy is the energy to minum block equipped with strain gauges is used as a excavate a unit volume of rock (kWh/m3) and s/d is dynamometer to record thrust forces up to 50 t. A data cutter spacing/cutting depth ratio. This phenomenon is illustrated in Figure 11 (Ozdemir, 1992). The first and second cuts are inefficient, optimum chip is formed 160 in the third pass given a minimum specific energy value which is preferred for an efficient excavation. 140 Groove deeping should be always avoided for suc- 120 cessful operations. Bearing in mind that the distance 100 between disc cutters in a TBM cutting head is con- 80 stant, the optimum s/d ratio will be directly dictate opti- 60 mum cutting depth. The relationship between depth 18.09.2000 FT (kN/disc) 40 of cut, disc thrust force, disc rolling force, may be 20 11.10.2000 obtained directly from full scale rock cutting tests real- ized on big size rock blocks representing the rock for- 0 mations which are likely to be found in tunnel line. 0312 456Thrust force for a predetermined depth of cut will Depth of cut (mm/rev) directly dictate total machine thrust which the machine operator must apply for an efficient cutting operation. Figure 9. The relationship between disc thrust force and Rolling force for a given disc cutting depth, may be depth of cut measured in the field. directly used to calculate the power consumed during the cutting process. The torque and the power consumed in optimum cutting conditions may be calculated with the following Equations (Bilgin, 1999).

(1)

(2)

In these Equations FR is rolling force, n the total number of disc cutters in cutting head of the TBM, ri is the distance from the cutter to the center of the machine, T is torque and N is revolution per minute of the cutting head and P is the power consumed during the cutting operation. Instantaneous cutting rate or net cutting rate may be calculated using the relation given Figure 10. Schematic view of full scale cutting rig. in Equation 3.

824 a compressive strength of 100 8MPa is given in (3) Figure 12. A disc cutter taken from the TBM was used in rock cutting tests. Disc has a diameter of 305 mm and edge width of 10 mm. The actual cutter forces and In the Equation 3 (ICR) is instantaneous cutting measured values of thrust forces are given in Table 5. 3 rate in m /h, k is the energy transfer ratio changing As seen from this table the field values are very close between 0.7 and 0.8, (P) is power consumed during to laboratory results meaning that thrust force values the cutting process and (SE) is optimum specific obtained in the laboratory may lead to predict field energy obtained in full scale laboratory cutting tests depth of cut values for 1 rev./mm, hence net cutting as explained in Figure 11. rate of TBM. The relationship between depth of cut and cutter In situ FT values for one disc are calculated by forces obtained in the laboratory for a limestone having dividing total thrust force applied by thrust cylinders of TBM divided by total disc numbers. However First pass actual thrust values delivered to rock surface is usu- ally less then calculated values due to system efficiency and losses such as back-up system towage during mining or friction losses from contact between the shield and the rock. This is experienced to be d1 around 20% during the excavation of Tarabya tunnel. It is reported by different investigators that net average cutter thrust force can be easily 40% less than Second pass, groove deeping average gross force calculated from hydraulic cylin- der pressures (Nelson, 1993). The optimum specific energy for the limestone rock formation excavated in Tarabya tunnel is calculated to be 5.7 kwh/3 for s/d ratio of 14. Cutter spacing of the face cutters in TBM is 86 mm, given an optimum d2 ridge forming 140 s 120 100 Third cut 80 60

d3 FN,FR (kN) 40 20 FR Optimum chip formed FN 0 d1 = first pass, inefficient depth of cut 061024 8 d2 = second pass, groove deepening Depth of Cut (mm) d3 = third pass, optimum chip formed Figure 12. The relationship between depth of cut and cutter forces obtained in the laboratory.

Table 5. Field and measured thrust force values.

In situ FT with 20% Measured Measured less due to laboratory Specific Energy Depth of cut in situ FT friction values optimum s/d: 10 to 20 (mm) (kN) factors, (kN) FT, (kN)

2 100 81 80 s/d 3 110 87 88 4 120 94 96 Figure 11. The relation between specific energy and opti- 5 135 101 108 mum cutter spacing/depth of cut.

825 depth of cut of 6 mm. Rolling force FR for one disc limestone, sandstone – siltstone and andasite dykes. for 6 mm depth of cut is calculated to be 9.82 kN from The mean monthly advance rate of 377 m was obtained Figure 12. The torque and the cutting power for opti- with a machine utilization time of 35%. It is shown that mum depth of cut are calculated using equation 1 and 2 laboratory full-scale cutting tests may serve as a use- as 142.4 kNm and 239 kW consecutively. In optimum ful guide for an efficient excavation. It is shown that cutting condition instantaneous cutting rate of the TBM optimum cutting conditions were obtained using a total may be calculated using equation 3 as given below. machine thrust force of 259 t given an optimum disc penetration of 6 mm/revolution.

7ACKNOWLEDGEMENT

The authors are grateful to Tinsa – Öztas¸ – Hazinedaroglu – Simelko consortium given access to Optimum thrust force for one disc cutter from field data and Prof.Dr. Levent Ozdemir from CSM, Figure 12 is found to be 10.8 t Optimum thrust force USA for helping to construct laboratory facilities. that TBM operator should apply taking into account the 20% friction losses; REFERENCES

Bilgin N. et al. 1999. The Performance Prediction of a TBM In field, during the excavation of Tarabya tunnel in Tuzla-Dragos Sewerage Tunnel. In: T.Allen et al. (eds). The World Tunnel Congress, Oslo 31 May 3 June: the instantaneous cutting rate of 33.5 m/h was obtained 817–822, Rotterdam: Balkema. during some shifts. Bilgin N. et al. 1999. The Performance Prediction of a TBM in Difficult Ground Condition. AFTES-Journees d’Etudes Internationales de Paris, 25–28 October, 115–121. Nelson P. 1993. TBM Performance Analysis with Reference 6 CONCLUSION to Rock Properties, Mechanized Excavation. In: J.A. Hudson (ed). Comprehensive Rock Engineering 4: 27. Tunnel excavation in Tarabya started in July 2000 and Ozdemir L. 1992. Short Course Notes. Mechanical Excava- finished in November 2004. The main rock formation tion Techniques in Underground Construction, 2 Volumes, excavated were Paleozoic aged sedimentary rocks, ITU Faculty of Mines.

826